Last month, news broke that owners of the New Orleans Times-Picayune are planning a major restructuring of that publication. The message arrived in Ann Arbor with an eerie familiarity. The same folks owned the former Ann Arbor News, a newspaper they closed in order to create a new company called AnnArbor.com.
The familiar part of the news includes severe staff reductions at the Times-Picayune and a shift in focus to online delivery, cutting back its printed edition to three days a week.
David Carr of the New York Times reported that changes at the Times-Picayune apparently would be modeled after the transformation in Ann Arbor. The Newhouse family – whose media holdings include the publications in Ann Arbor and New Orleans, among dozens of others nationwide – had made Ann Arbor its testbed for this approach in 2009.
Residents of New Orleans have my deepest sympathies.
The decisions about the Times-Picayune are disturbing, even if considered independently of other Newhouse operations. But especially disturbing is the idea that AnnArbor.com might serve as a model for anything.
The news from New Orleans coincided with an ultimately successful effort by The Ann Arbor Chronicle to push AnnArbor.com to correct a shockingly flawed analysis related to fire protection that had been originally reported by Ryan Stanton back in May of 2011. Within days of publication last year, Chronicle editor Dave Askins alerted Stanton to the likely source of the factual errors in Stanton’s piece.
Askins correctly analyzed the Ann Arbor fire department’s reports that Stanton had misinterpreted, and soon after that The Chronicle published that analysis. It wasn’t until this week, though, that AnnArbor.com’s “chief content officer,” Tony Dearing, wrote a column acknowledging the fact that the response times reported by Stanton were inaccurate. But Dearing’s accounting of AnnArbor.com’s errors is misleading and incomplete – in part because it fails to take responsibility for obvious reporting mistakes, blaming sources instead.
In that respect, Dearing’s column continues a pattern of disingenuous communication by AnnArbor.com with the community it purports to serve.
I realize there’s a certain etiquette I’m violating in calling out the leadership of another publication in this way. What I hear on a regular basis about the community’s perception of the quality of reporting and editorial oversight at AnnArbor.com ranges from idle snark to complete outrage. But our Midwestern culture exerts a firm pressure to make nice and get along. And for some community members, a certain fatigue has set in, along with a sense that it’s not worth the energy to rehash these things – it’s time to move on. To some extent I actually agree with that. It would be nice to move on.
But a polite culture and need to look forward do not justify turning away from some real problems with AnnArbor.com’s basic approach to community service. That’s especially true as the Newhouses roll out the Ann Arbor model in other markets.
What’s more, given the marketing resources of AnnArbor.com’s New York-based owners, there’s a risk that a funhouse-mirror version of reality will become accepted as accurate, and could inappropriately influence public policy in a way that causes long-term damage to this community. That’s unacceptable.
In this column, I’ll explain how the fire protection saga unfolded, what it reflects about AnnArbor.com and the state of traditional media, and the importance of being grounded in the community you cover.
Fire Safety: A Story of Flawed Reporting
Before I launch into the fire department response time analysis, let me acknowledge that not every reader will have the stomach for this level of detail. If you’d rather not read about “notify times” and “en route times,” or what it finally took to convince AnnArbor.com that its initial reporting might have been inaccurate, then skip to the next section.
The story, which Dearing finally acknowledged last week was in error, was written by Ryan Stanton and published in May of last year – just before the Ann Arbor city council considered an annual budget that called for a reduction in firefighter positions. The story served the basic editorial stance of AnnArbor.com: By decreasing fire department budget resources, the Ann Arbor city council was impeding firefighters’ ability to cover the distance between their stations and the scenes of major fires in a timely fashion.
To support that narrative, Stanton presented his readers with travel times for four major fires that he claimed exceeded the national travel time standard of four minutes for a first-arriving company.
Certainly, if a fire department typically records travel times that exceed the national standard, it indicates that the number of staffed fire stations in a geographic area is not sufficient. So the travel time is an appropriate place to focus for an investigative enterprise that seeks to answer the question: Are fire department resources adequate?
To start with a general observation, Stanton’s story was unfortunately vague with respect to its terminology – using “response time” instead of “travel time.” He did not explain explicitly to readers that “response time,” as used throughout his story, was meant specifically to refer to the “travel time” – the time interval from the station to the fire scene. But given the story’s use of the “travel time” standard of four minutes, it’s evident that Stanton’s use of “response time” throughout his piece is, in fact, a reference to travel time.
At a city council meeting, a day after publication of that story, Barnett Jones – who then served as the city’s chief of safety services – publicly called out Stanton for mistakes in the story, including inaccurately-calculated response times. In a scolding email that Stanton subsequently sent to Jones, justifying his story’s report of response times, it’s also clear that the intent of Stanton’s story was to present travel times to readers.
Tony Dearing also admitted in a May 2, 2012 meeting with Chronicle editor Dave Askins that the intent of Stanton’s story was to calculate and present travel times to readers. That’s a point of common ground, actually – the idea that a relevant data point for measuring the adequacy of a fire department’s resources is the “travel time.”
After hearing Jones’ remarks at the May 16, 2011 city council meeting, that same evening Askins gave a cursory review to the records that Stanton used to write his story. He then emailed Stanton, also that same evening, pointing out to Stanton the likely source of his error.
To understand the significance of that emailed message, it’s important first to understand the difference between a time point – like “en route time” or “notify time” – and a time interval, like “travel time.” It’s time points, not intervals, that are recorded in fire department records. To get an interval from the fire department records, you have to do a calculation – in this case, starting from the time point recorded as “arrival time” – the time a fire truck arrived on the scene.
How do you calculate the travel time interval? The plain language of the National Fire Protection Association standards would lead a reader to conclude that it’s the “en route time” that should be subtracted from “arrival time” to calculate “travel time.” But in his email to Jones, Stanton justified the same conclusion in a different way – by citing an unnamed authority from Massachusetts: “I calculated the response times based on how an NFPA representative in Massachusetts told me I should calculate them, which is to clock the 4-minute travel time starting from the first vehicle’s ‘en route’ time.”
And to be fair to Stanton, all other things being equal, you should be able to look at some city’s fire department reports, pick out the “arrival time” and the “en route time,” perform the clock arithmetic and get an accurate travel time. Of course, that assumes the “en route times” in the reports are accurate.
But even at first glance, it’s evident that the Ann Arbor fire department reports show “en route times” that are likely inaccurate. That’s because they’re recorded as identical, down to the second, with another time point recorded as “notify time” – the time the alarm was given.
From the fire department reports, it’s not hard to reach at least a tentative conclusion that the fire department is only interested in the sum of the two intervals, which would be possible to calculate if it’s the “notify time” that’s accurate. That’s what Askins pointed out to Stanton in the email he sent the same evening as the May 16, 2011 council meeting:
In the AAFD reports, the times recorded for “enroute time” and “notify time” are identical. That may be the source of the confusion. I’d guess that “notify time” is accurate and filled in from call-center information, and that “enroute time” is just systematically copied from “notify time” into that slot.
Stanton’s emailed reply to Askins blamed Stanton’s sources:
If that’s the case, their reports are wrong and they’re blaming me? They knew I was calculating response times, and they give me bad data? How unfortunate.
I won’t venture to speculate what the fire department knew about Stanton’s reporting or intentions. But it’s clear that before he wrote the story, Stanton did not ask anyone locally a question that yielded an accurate description of the information contained in Ann Arbor fire department reports. Instead, he asked a question of someone in a place called Massachusetts.
The right question to ask of the local Ann Arbor fire department would have been: Which one of these two time points is accurate, and what is the actual value of the other time point? That way, you could calculate a “turnout time” interval for the fires as well as a “travel time.” The “turnout time” is the interval between the alarm and the start of the travel time interval. That’s an important interval, because it measures how quickly firefighters can get into their gear and onto their trucks. But at their May 2, 2012 meeting, Dearing admitted to Askins that AnnArbor.com had not attempted to calculate a “turnout time” for any of the fires.
For one of the fires that AnnArbor.com tried to analyze, the answer was already included in a screen shot taken from the computer-aided dispatch (CAD) screen, which was among the materials provided by AAFD to Stanton for his story. That screenshot showed different times for the CAD analogs of “notify time” and “en route time” – which the AAFD reports systematically show as identical. But Stanton apparently did not incorporate the information from that screen shot into his reporting.
Not long after initial publication of Stanton’s article, AnnArbor.com corrected the time for the one fire that had included a CAD screen shot. The note of correction, however, claimed it was “based on new information provided to AnnArbor.com.” When Askins confronted Dearing at their May 2, 2012 meeting about the disingenuousness of calling the CAD screen shot “new information,” Dearing insisted that AAFD had provided a new, corrected report to AnnArbor.com – and that AnnArbor.com’s correction was based only on that new AAFD report. When Askins pointed out that the corrected “travel time” in Stanton’s story matched exactly the time indicated in the CAD screenshot, Dearing admitted that it did – but he still insisted that the correction was based only on the “new information.”
Calling that a case of “new information” masks Stanton’s failure to notice existing information, and it’s a misleading accounting of the admitted error. But it gets worse, partly because that’s where AnnArbor.com staff apparently stopped working as reporters. If you got one fire wrong, what about the three others?
It was only after a series of emails and a voicemail to Dearing that he finally agreed recently to meet with Askins. Among the facts that Askins had suggested Dearing review with Stanton was basic geographic information about the fires. In fact, that geographic information was included in a spreadsheet that Askins recently sent directly to Dearing, after publishing a link to it several months earlier as part of two different Chronicle reports. The spreadsheet contains CAD data for all four fires that AnnArbor.com tried to analyze – data The Chronicle was able to obtain through unofficial channels, because we continued to report on this topic. That CAD data allowed us to have confidence in our published conclusion about the four fires:
The time interval that seemed much longer than it should be (based on national standards), and that provides the greatest opportunity for improvement is the interval between the time a call comes to a station and the time a firetruck starts rolling to the scene (turnout time). That is to say, the other time interval – the travel time from fire station to fire scene – did not look like the place where the AAFD could improve most.
Why did Askins want Dearing to take a look at geography here? It’s not just because a “sense of place” is generally important for journalists serving a community. It’s because when it comes to calculating travel times, it’s an obvious question to ask: What was the distance traveled, and is it plausible that a fire engine would take that long to get there?
Yet at their meeting, Dearing told Askins that Stanton’s initial reporting had not considered geography.
Dearing also admitted that he himself had up to that point not considered the travel distance – even while claiming that he’d reviewed everything in detail, both with Stanton and with fire chief Chuck Hubbard.
Only when Askins showed Dearing where the fire scenes were on a map in relation to the fire stations did Dearing finally appear to take seriously the possibility that he and Stanton had been wrong about travel times all along.
At their recent meeting, Dearing finally admitted the obvious to Askins – that no, it was not really plausible that a fire truck would take 4 minutes and 9 seconds to travel roughly half a mile in the middle of the night. Also not plausible is that a fire truck would take 6 minutes and 15 seconds to travel roughly 1.1 miles, in the early morning hours when no traffic would be anticipated. Dearing told Askins: “You’ve given me more work to do.” It’s work Askins had already done.
To make any claim of being honest and forthright with his readers, Dearing’s column admitting the errors needed to include the geography of those fire responses – but Dearing’s column is silent on that subject. Including the geography would have made it clear not just that AnnArbor.com got the travel times wrong, but that the reporter and editor had no one to blame but themselves for getting those times wrong.
Instead, what Dearing’s column offered his readers was the same kind of disingenuous explanation that AnnArbor.com published with the initial correction of Stanton’s story – that there was “additional information” and that they “were told” something that turned out not to be accurate. From Dearing’s column:
Since AnnArbor.com first reported last year that the department was struggling to meet response time standards, a great deal of additional information is now available, and based on that information, we owe the community a more complete and accurate analysis of this issue than we have offered to this point. … Our original reporting was based on reports supplied to us by the city, which listed en route times. We were told that en route times represented travel times.
This “additional information” was already available – and that’s why The Chronicle had already published it, starting with the city council meeting report published within days of Stanton’s initial story, followed with later analyses.
An Aside: Some Thoughts About Awards
It’s worth noting that Stanton’s article about fire safety won a first-place award from the Michigan Associated Press for investigative reporting. And yes, I spewed my coffee when I heard about that.
Though the Pulitzers might be the most notable exception, journalism awards can be a rather incestuous affair. For the Michigan Associated Press, for example, only publications that pay to be members of the AP are eligible for the awards.
I speak here from my experience as a contest judge during my tenure at the Ann Arbor News. In many contests, submissions are shipped off to judges in another market for review. The judges are typically overworked editors who have scant time to spare on this task.
It’s difficult to get a sense of an article that lands in your lap without context. There might be a cover letter with some explanation provided, but of course those are submitted by the organization hoping to win an award. Frankly, in many cases there’s little to distinguish one entry from another – and I’m sure that was the case for many of the awards that my colleagues and I won while working at The News.
Certainly there’s no time or inclination to vet the award submissions for accuracy, though it’s typically required that any correction made on an article should be noted in the submission materials. So the process relies on the integrity of each publication to be forthright about the quality of its submissions.
In this case, given that Dearing’s column outlining problems with the analysis wasn’t published until well after the awards were handed out, there’s no doubt that the judges were unaware of that full context.
We’ve asked AP’s regional bureau chief if Michigan AP will be reviewing its award to the fire response story, but haven’t received a reply. I’m not holding my breath – AnnArbor.com is a member, and The Ann Arbor Chronicle is not.
New Model of Doing Business?
I should pause here to note that my criticism of AnnArbor.com is not based on some self-righteous belief that if a mistake is made it must be because the reporter wasn’t conscientious. It’s not possible to do this job – or any job – without error. Even the most meticulous, conscientious reporter will screw up from time to time. We make our own share of mistakes. Corrected Chronicle errors are easy to spot in the text – because “deleted material” is denoted with red strike-through text and added material is denoted with blue text. The result is not pretty. Believe me, it’s not fun to make such an ostentatious accounting of our mistakes, but we do.
AnnArbor.com’s approach to errors is different. Dearing has simply added a note to the top of Stanton’s article. The errors that remain in the text are apparently left to readers to sort out for themselves.
AnnArbor.com’s approach to correcting Stanton’s story is part of an ongoing pattern – failing to be forthright with the community. It’s a pattern that I’ve noted previously.
In a March 13, 2011 column “History Repeats at AnnArbor.com,” I described how some news about a round of layoffs at AnnArbor.com had not been shared with the Ann Arbor community. The layoffs were eventually acknowledged, after a reader posted a question about the dismissals a few days after the fact, on a section of the AnnArbor.com website called the Community Wall. The response was a two-paragraph comment from Dearing that started off with the corporate-speak of “personnel issues are an internal matter and we don’t discuss them publicly…” He continued by acknowledging that ”I can confirm that we reorganized our newsroom this week to put our focus more squarely on local news coverage.”
As I wrote at the time, his explanation was insulting – who on earth would view cuts to local reporting staff as a way to focus on local news coverage? It was also evocative of a column written by former Ann Arbor News editor Ed Petykiewicz in December 2008, a few months after we launched The Chronicle. In the wake of buyouts at the newspaper and a shrinking staff, Petykiewicz claimed that the newspaper would be focusing more on local content – and just four months later, the announcement came that the News would close. Perhaps Dearing and Petykiewicz were both looking at the world through a common Newhouse/funhouse mirror of reality – I don’t know.
The misrepresentation of basic reality is shown on the business side as well. When the Newhouse family closed the Ann Arbor News in 2009, the narrative relied crucially on the idea that the newly-formed business to replace the News was a “startup” like any other startup. No one in the community really bought that story, so there was no surprise or objection when the AnnArbor.com executive leadership subsequently accepted an award from the Ann Arbor Ypsilanti Regional Chamber of Commerce, honoring businesses that had been members for several decades.
An early marketing campaign for AnnArbor.com also tried to highlight longevity. Responding to criticism about the inexperience of their reporting staff – because many of the senior editors and reporters at the Ann Arbor News were not rehired by AnnArbor.com – the company took out billboards trumpeting the collective experience in journalism of its entire staff, measured by adding up each employee’s years of experience. Since then, turnover has been frequent. Many of the original staff are no longer there, including most of the initial key hires, and the experience level of reporters has continued to drop.
That means even fewer people remain at AnnArbor.com who have deep connections to the community, with a sense of history and context. But here’s the thing: It’s easier to operate under those conditions when these qualities don’t really matter.
What does matter in the age of “churnalism” is the ability to quickly push out spot news, rewrites of press releases, rewrites of other publication’s articles, “instant analysis” – an oxymoron if there ever was one – and other fodder to drive site traffic, and in turn generate ad revenue.
Highlighting drama and conflict has always been a staple of mainstream media, and provocative, misleading headlines are nothing new. Some readers of AnnArbor.com like to grouse about the bottom-feeding nature of the comments left on articles, but in many cases the comments seem like simply an amped-up version of the stories themselves. When a publication trades on fomenting artificial controversy, is it really a surprise when the comments written there reflect the community’s lowest common denominator?
That kind of storytelling approach to journalism, which relies on identifying characters in conflict, comes with inherent dangers. In last month’s Chronicle milestone column, Dave Askins laid out the perils of that approach, and contrasted it with The Chronicle’s emphasis on description, analysis and explanation.
There are many problems with storytelling as a way to convey news and information, but chief among them is that the reader must rely on the integrity of the storyteller, because facts don’t play a prominent role. You rely on the writer having a deep understanding of the topic, its history and context – and a strong sense of place. Absent that integrity, all you’re left with is a hollow collection of words.
Implications for Community
So should AnnArbor.com be the “model” for the future of the New Orleans Times-Picayune? From my perspective and for much of the community, the AnnArbor.com model isn’t working well in Ann Arbor. I conveyed that sentiment to Steve Myers, managing editor of Poynter.org, for an article he wrote following the Times-Picayune announcement.
Ann Arbor is a community that’s relatively small, relatively wealthy, highly educated, with a high percentage of people who have access to the Internet. For those who don’t, there’s a strong library and school system to help pick up the slack.
These same conditions don’t exist in New Orleans.
A book has been circulating among local business and government leaders called “Economics of Place,” published by the Ann Arbor-based Michigan Municipal League. The ideas in it aren’t new, but they’ve been packaged in a way that seems to resonate with people who are looking to articulate what they like about where they live.
A newspaper – online or printed – can play a crucial role in reflecting and bolstering that sense of place, and in leading the community to an even better version of itself. But it can’t do that with a superficial, false understanding of the community it serves, or by misleading readers.
That’s true in Ann Arbor, in New Orleans – and anywhere else.
Mary Morgan is publisher and co-owner of The Ann Arbor Chronicle. The monthly milestone column, which appears on the second day of each month – the anniversary of The Chronicle’s Sept. 2, 2008 launch – is an opportunity for either the publisher or the editor of The Chronicle to touch base with readers on topics related to this publication.